AlekseyKorshuk commited on
Commit
d87ec48
1 Parent(s): 17a4171

huggingartists

Browse files
README.md ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ datasets:
4
+ - huggingartists/lady-gaga
5
+ tags:
6
+ - huggingartists
7
+ - lyrics
8
+ - lm-head
9
+ - causal-lm
10
+ widget:
11
+ - text: "I am"
12
+ ---
13
+
14
+ <div class="inline-flex flex-col" style="line-height: 1.5;">
15
+ <div class="flex">
16
+ <div
17
+ style="display:DISPLAY_1; margin-left: auto; margin-right: auto; width: 92px; height:92px; border-radius: 50%; background-size: cover; background-image: url(&#39;https://images.genius.com/e7e76c378cb43b4b1ff03947d5c0481a.400x400x1.jpg&#39;)">
18
+ </div>
19
+ </div>
20
+ <div style="text-align: center; margin-top: 3px; font-size: 16px; font-weight: 800">🤖 HuggingArtists Model 🤖</div>
21
+ <div style="text-align: center; font-size: 16px; font-weight: 800">Lady Gaga</div>
22
+ <a href="https://genius.com/artists/lady-gaga">
23
+ <div style="text-align: center; font-size: 14px;">@lady-gaga</div>
24
+ </a>
25
+ </div>
26
+
27
+ I was made with [huggingartists](https://github.com/AlekseyKorshuk/huggingartists).
28
+
29
+ Create your own bot based on your favorite artist with [the demo](https://colab.research.google.com/github/AlekseyKorshuk/huggingartists/blob/master/huggingartists-demo.ipynb)!
30
+
31
+ ## How does it work?
32
+
33
+ To understand how the model was developed, check the [W&B report](https://wandb.ai/huggingartists/huggingartists/reportlist).
34
+
35
+ ## Training data
36
+
37
+ The model was trained on lyrics from Lady Gaga.
38
+
39
+ Dataset is available [here](https://huggingface.co/datasets/huggingartists/lady-gaga).
40
+ And can be used with:
41
+
42
+ ```python
43
+ from datasets import load_dataset
44
+
45
+ dataset = load_dataset("huggingartists/lady-gaga")
46
+ ```
47
+
48
+ [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/s6109nue/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
+
50
+ ## Training procedure
51
+
52
+ The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Lady Gaga's lyrics.
53
+
54
+ Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/3g2fq2s0) for full transparency and reproducibility.
55
+
56
+ At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/3g2fq2s0/artifacts) is logged and versioned.
57
+
58
+ ## How to use
59
+
60
+ You can use this model directly with a pipeline for text generation:
61
+
62
+ ```python
63
+ from transformers import pipeline
64
+ generator = pipeline('text-generation',
65
+ model='huggingartists/lady-gaga')
66
+ generator("I am", num_return_sequences=5)
67
+ ```
68
+
69
+ Or with Transformers library:
70
+
71
+ ```python
72
+ from transformers import AutoTokenizer, AutoModelWithLMHead
73
+
74
+ tokenizer = AutoTokenizer.from_pretrained("huggingartists/lady-gaga")
75
+
76
+ model = AutoModelWithLMHead.from_pretrained("huggingartists/lady-gaga")
77
+ ```
78
+
79
+ ## Limitations and bias
80
+
81
+ The model suffers from [the same limitations and bias as GPT-2](https://huggingface.co/gpt2#limitations-and-bias).
82
+
83
+ In addition, the data present in the user's tweets further affects the text generated by the model.
84
+
85
+ ## About
86
+
87
+ *Built by Aleksey Korshuk*
88
+
89
+ [![Follow](https://img.shields.io/github/followers/AlekseyKorshuk?style=social)](https://github.com/AlekseyKorshuk)
90
+
91
+ [![Follow](https://img.shields.io/twitter/follow/alekseykorshuk?style=social)](https://twitter.com/intent/follow?screen_name=alekseykorshuk)
92
+
93
+ [![Follow](https://img.shields.io/badge/dynamic/json?color=blue&label=Telegram%20Channel&query=%24.result&url=https%3A%2F%2Fapi.telegram.org%2Fbot1929545866%3AAAFGhV-KKnegEcLiyYJxsc4zV6C-bdPEBtQ%2FgetChatMemberCount%3Fchat_id%3D-1001253621662&style=social&logo=telegram)](https://t.me/joinchat/_CQ04KjcJ-4yZTky)
94
+
95
+ For more details, visit the project repository.
96
+
97
+ [![GitHub stars](https://img.shields.io/github/stars/AlekseyKorshuk/huggingartists?style=social)](https://github.com/AlekseyKorshuk/huggingartists)
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "gpt2",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2LMHeadModel"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 50256,
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 50256,
11
+ "gradient_checkpointing": false,
12
+ "initializer_range": 0.02,
13
+ "layer_norm_epsilon": 1e-05,
14
+ "model_type": "gpt2",
15
+ "n_ctx": 1024,
16
+ "n_embd": 768,
17
+ "n_head": 12,
18
+ "n_inner": null,
19
+ "n_layer": 12,
20
+ "n_positions": 1024,
21
+ "resid_pdrop": 0.1,
22
+ "scale_attn_weights": true,
23
+ "summary_activation": null,
24
+ "summary_first_dropout": 0.1,
25
+ "summary_proj_to_labels": true,
26
+ "summary_type": "cls_index",
27
+ "summary_use_proj": true,
28
+ "task_specific_params": {
29
+ "text-generation": {
30
+ "do_sample": true,
31
+ "max_length": 200,
32
+ "min_length": 100,
33
+ "temperature": 1.0,
34
+ "top_p": 0.95
35
+ }
36
+ },
37
+ "torch_dtype": "float32",
38
+ "transformers_version": "4.10.0",
39
+ "use_cache": true,
40
+ "vocab_size": 50257
41
+ }
evaluation.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eval_loss": 1.7604948282241821, "eval_runtime": 8.1984, "eval_samples_per_second": 22.078, "eval_steps_per_second": 2.805, "epoch": 20.0}
flax_model.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7afa87885c4237efdbd54f14a90c394a7789cee5d9dc805cefa09922257eca9
3
+ size 497764120
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e975e394301ef8deef76b9ee8c42cf177aa52fc2de166eb69e8a52fa5f03089
3
+ size 995604017
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:075245bdc5c5f7cd7a9cdac5907183e6b2f805ffe1d84da4d6b08b01b13cdde7
3
+ size 510403817
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03a2d5e765e5cf5ecb55bf1d575b6d6e34d62c85ee9fecda0b266d1418fa375b
3
+ size 14503
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e3679dd478a770c99d8cf2c99947149cd3de03554a50244d19b822d7b456455
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,1048 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.7604948282241821,
3
+ "best_model_checkpoint": "output/lady-gaga/checkpoint-822",
4
+ "epoch": 6.0,
5
+ "global_step": 822,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.04,
12
+ "learning_rate": 0.0001367495807816491,
13
+ "loss": 2.9044,
14
+ "step": 5
15
+ },
16
+ {
17
+ "epoch": 0.07,
18
+ "learning_rate": 0.000135404237921706,
19
+ "loss": 2.7308,
20
+ "step": 10
21
+ },
22
+ {
23
+ "epoch": 0.11,
24
+ "learning_rate": 0.0001331816381338556,
25
+ "loss": 2.5335,
26
+ "step": 15
27
+ },
28
+ {
29
+ "epoch": 0.15,
30
+ "learning_rate": 0.00013011096805539083,
31
+ "loss": 2.6599,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.18,
36
+ "learning_rate": 0.00012623255097542413,
37
+ "loss": 2.6824,
38
+ "step": 25
39
+ },
40
+ {
41
+ "epoch": 0.22,
42
+ "learning_rate": 0.00012159731731930818,
43
+ "loss": 2.5916,
44
+ "step": 30
45
+ },
46
+ {
47
+ "epoch": 0.26,
48
+ "learning_rate": 0.00011626613584273574,
49
+ "loss": 2.4845,
50
+ "step": 35
51
+ },
52
+ {
53
+ "epoch": 0.29,
54
+ "learning_rate": 0.00011030901431811399,
55
+ "loss": 2.4155,
56
+ "step": 40
57
+ },
58
+ {
59
+ "epoch": 0.33,
60
+ "learning_rate": 0.00010380418020960441,
61
+ "loss": 2.386,
62
+ "step": 45
63
+ },
64
+ {
65
+ "epoch": 0.36,
66
+ "learning_rate": 9.683705340917955e-05,
67
+ "loss": 2.4477,
68
+ "step": 50
69
+ },
70
+ {
71
+ "epoch": 0.4,
72
+ "learning_rate": 8.949912452347496e-05,
73
+ "loss": 2.2686,
74
+ "step": 55
75
+ },
76
+ {
77
+ "epoch": 0.44,
78
+ "learning_rate": 8.188675344149976e-05,
79
+ "loss": 2.4056,
80
+ "step": 60
81
+ },
82
+ {
83
+ "epoch": 0.47,
84
+ "learning_rate": 7.409990396012103e-05,
85
+ "loss": 2.4163,
86
+ "step": 65
87
+ },
88
+ {
89
+ "epoch": 0.51,
90
+ "learning_rate": 6.62408310839107e-05,
91
+ "loss": 2.239,
92
+ "step": 70
93
+ },
94
+ {
95
+ "epoch": 0.55,
96
+ "learning_rate": 5.841273823741302e-05,
97
+ "loss": 2.3147,
98
+ "step": 75
99
+ },
100
+ {
101
+ "epoch": 0.58,
102
+ "learning_rate": 5.071842202299234e-05,
103
+ "loss": 2.0396,
104
+ "step": 80
105
+ },
106
+ {
107
+ "epoch": 0.62,
108
+ "learning_rate": 4.3258922320970566e-05,
109
+ "loss": 2.2021,
110
+ "step": 85
111
+ },
112
+ {
113
+ "epoch": 0.66,
114
+ "learning_rate": 3.613219545860942e-05,
115
+ "loss": 2.3175,
116
+ "step": 90
117
+ },
118
+ {
119
+ "epoch": 0.69,
120
+ "learning_rate": 2.9431827871557342e-05,
121
+ "loss": 2.2247,
122
+ "step": 95
123
+ },
124
+ {
125
+ "epoch": 0.73,
126
+ "learning_rate": 2.3245807149646032e-05,
127
+ "loss": 2.3038,
128
+ "step": 100
129
+ },
130
+ {
131
+ "epoch": 0.77,
132
+ "learning_rate": 1.7655366605361312e-05,
133
+ "loss": 2.2539,
134
+ "step": 105
135
+ },
136
+ {
137
+ "epoch": 0.8,
138
+ "learning_rate": 1.2733918537834174e-05,
139
+ "loss": 2.4522,
140
+ "step": 110
141
+ },
142
+ {
143
+ "epoch": 0.84,
144
+ "learning_rate": 8.546090200467903e-06,
145
+ "loss": 2.1125,
146
+ "step": 115
147
+ },
148
+ {
149
+ "epoch": 0.88,
150
+ "learning_rate": 5.146875131638653e-06,
151
+ "loss": 2.1399,
152
+ "step": 120
153
+ },
154
+ {
155
+ "epoch": 0.91,
156
+ "learning_rate": 2.5809109929881027e-06,
157
+ "loss": 2.159,
158
+ "step": 125
159
+ },
160
+ {
161
+ "epoch": 0.95,
162
+ "learning_rate": 8.818933985584571e-07,
163
+ "loss": 2.4269,
164
+ "step": 130
165
+ },
166
+ {
167
+ "epoch": 0.99,
168
+ "learning_rate": 7.213343222279864e-08,
169
+ "loss": 2.4774,
170
+ "step": 135
171
+ },
172
+ {
173
+ "epoch": 1.0,
174
+ "eval_loss": 2.1954123973846436,
175
+ "eval_runtime": 8.0612,
176
+ "eval_samples_per_second": 22.453,
177
+ "eval_steps_per_second": 2.853,
178
+ "step": 137
179
+ },
180
+ {
181
+ "epoch": 1.02,
182
+ "learning_rate": 1.6226466398373308e-07,
183
+ "loss": 1.9999,
184
+ "step": 140
185
+ },
186
+ {
187
+ "epoch": 1.06,
188
+ "learning_rate": 1.1511035125304852e-06,
189
+ "loss": 1.9351,
190
+ "step": 145
191
+ },
192
+ {
193
+ "epoch": 1.09,
194
+ "learning_rate": 3.0256647877395575e-06,
195
+ "loss": 2.077,
196
+ "step": 150
197
+ },
198
+ {
199
+ "epoch": 1.13,
200
+ "learning_rate": 5.7613322090181854e-06,
201
+ "loss": 2.0647,
202
+ "step": 155
203
+ },
204
+ {
205
+ "epoch": 1.17,
206
+ "learning_rate": 9.322181660285877e-06,
207
+ "loss": 2.0654,
208
+ "step": 160
209
+ },
210
+ {
211
+ "epoch": 1.2,
212
+ "learning_rate": 1.3661452936688322e-05,
213
+ "loss": 2.1723,
214
+ "step": 165
215
+ },
216
+ {
217
+ "epoch": 1.24,
218
+ "learning_rate": 1.8722163788180865e-05,
219
+ "loss": 2.0249,
220
+ "step": 170
221
+ },
222
+ {
223
+ "epoch": 1.28,
224
+ "learning_rate": 2.4437858196511952e-05,
225
+ "loss": 2.3832,
226
+ "step": 175
227
+ },
228
+ {
229
+ "epoch": 1.31,
230
+ "learning_rate": 3.0733479059415244e-05,
231
+ "loss": 2.0252,
232
+ "step": 180
233
+ },
234
+ {
235
+ "epoch": 1.35,
236
+ "learning_rate": 3.752635382213567e-05,
237
+ "loss": 2.119,
238
+ "step": 185
239
+ },
240
+ {
241
+ "epoch": 1.39,
242
+ "learning_rate": 4.472728011321611e-05,
243
+ "loss": 2.1383,
244
+ "step": 190
245
+ },
246
+ {
247
+ "epoch": 1.42,
248
+ "learning_rate": 5.224169712824147e-05,
249
+ "loss": 2.3061,
250
+ "step": 195
251
+ },
252
+ {
253
+ "epoch": 1.46,
254
+ "learning_rate": 5.997092737921572e-05,
255
+ "loss": 1.9195,
256
+ "step": 200
257
+ },
258
+ {
259
+ "epoch": 1.5,
260
+ "learning_rate": 6.781347250322049e-05,
261
+ "loss": 2.3574,
262
+ "step": 205
263
+ },
264
+ {
265
+ "epoch": 1.53,
266
+ "learning_rate": 7.566634611411474e-05,
267
+ "loss": 1.9516,
268
+ "step": 210
269
+ },
270
+ {
271
+ "epoch": 1.57,
272
+ "learning_rate": 8.34264261945917e-05,
273
+ "loss": 2.1115,
274
+ "step": 215
275
+ },
276
+ {
277
+ "epoch": 1.61,
278
+ "learning_rate": 9.099180926931044e-05,
279
+ "loss": 2.2426,
280
+ "step": 220
281
+ },
282
+ {
283
+ "epoch": 1.64,
284
+ "learning_rate": 9.82631485764331e-05,
285
+ "loss": 2.1105,
286
+ "step": 225
287
+ },
288
+ {
289
+ "epoch": 1.68,
290
+ "learning_rate": 0.00010514495866502442,
291
+ "loss": 1.8663,
292
+ "step": 230
293
+ },
294
+ {
295
+ "epoch": 1.72,
296
+ "learning_rate": 0.0001115468692866605,
297
+ "loss": 1.8012,
298
+ "step": 235
299
+ },
300
+ {
301
+ "epoch": 1.75,
302
+ "learning_rate": 0.00011738481211544774,
303
+ "loss": 1.9638,
304
+ "step": 240
305
+ },
306
+ {
307
+ "epoch": 1.79,
308
+ "learning_rate": 0.00012258212471273538,
309
+ "loss": 1.8965,
310
+ "step": 245
311
+ },
312
+ {
313
+ "epoch": 1.82,
314
+ "learning_rate": 0.0001270705572395265,
315
+ "loss": 2.1566,
316
+ "step": 250
317
+ },
318
+ {
319
+ "epoch": 1.86,
320
+ "learning_rate": 0.00013079116869668585,
321
+ "loss": 1.7786,
322
+ "step": 255
323
+ },
324
+ {
325
+ "epoch": 1.9,
326
+ "learning_rate": 0.00013369510092373763,
327
+ "loss": 1.9581,
328
+ "step": 260
329
+ },
330
+ {
331
+ "epoch": 1.93,
332
+ "learning_rate": 0.00013574422019228965,
333
+ "loss": 1.8796,
334
+ "step": 265
335
+ },
336
+ {
337
+ "epoch": 1.97,
338
+ "learning_rate": 0.00013691161796883605,
339
+ "loss": 2.1161,
340
+ "step": 270
341
+ },
342
+ {
343
+ "epoch": 2.0,
344
+ "eval_loss": 2.0334055423736572,
345
+ "eval_runtime": 8.0619,
346
+ "eval_samples_per_second": 22.451,
347
+ "eval_steps_per_second": 2.853,
348
+ "step": 274
349
+ },
350
+ {
351
+ "epoch": 2.01,
352
+ "learning_rate": 0.00013718196427104403,
353
+ "loss": 1.8034,
354
+ "step": 275
355
+ },
356
+ {
357
+ "epoch": 2.04,
358
+ "learning_rate": 0.00013655170897733555,
359
+ "loss": 1.7696,
360
+ "step": 280
361
+ },
362
+ {
363
+ "epoch": 2.08,
364
+ "learning_rate": 0.00013502912844621666,
365
+ "loss": 1.9071,
366
+ "step": 285
367
+ },
368
+ {
369
+ "epoch": 2.12,
370
+ "learning_rate": 0.00013263421683315968,
371
+ "loss": 1.6146,
372
+ "step": 290
373
+ },
374
+ {
375
+ "epoch": 2.15,
376
+ "learning_rate": 0.00012939842353223955,
377
+ "loss": 1.6415,
378
+ "step": 295
379
+ },
380
+ {
381
+ "epoch": 2.19,
382
+ "learning_rate": 0.0001253642401903761,
383
+ "loss": 1.6821,
384
+ "step": 300
385
+ },
386
+ {
387
+ "epoch": 2.23,
388
+ "learning_rate": 0.0001205846427174097,
389
+ "loss": 1.7192,
390
+ "step": 305
391
+ },
392
+ {
393
+ "epoch": 2.26,
394
+ "learning_rate": 0.0001151223956193987,
395
+ "loss": 1.8815,
396
+ "step": 310
397
+ },
398
+ {
399
+ "epoch": 2.3,
400
+ "learning_rate": 0.00010904922779046358,
401
+ "loss": 1.7244,
402
+ "step": 315
403
+ },
404
+ {
405
+ "epoch": 2.34,
406
+ "learning_rate": 0.00010244489058647663,
407
+ "loss": 1.9117,
408
+ "step": 320
409
+ },
410
+ {
411
+ "epoch": 2.37,
412
+ "learning_rate": 9.539611054974517e-05,
413
+ "loss": 1.7657,
414
+ "step": 325
415
+ },
416
+ {
417
+ "epoch": 2.41,
418
+ "learning_rate": 8.799545053725109e-05,
419
+ "loss": 1.8888,
420
+ "step": 330
421
+ },
422
+ {
423
+ "epoch": 2.45,
424
+ "learning_rate": 8.034009420783132e-05,
425
+ "loss": 1.617,
426
+ "step": 335
427
+ },
428
+ {
429
+ "epoch": 2.48,
430
+ "learning_rate": 7.253056983011824e-05,
431
+ "loss": 1.6686,
432
+ "step": 340
433
+ },
434
+ {
435
+ "epoch": 2.52,
436
+ "learning_rate": 6.46694301698818e-05,
437
+ "loss": 1.7576,
438
+ "step": 345
439
+ },
440
+ {
441
+ "epoch": 2.55,
442
+ "learning_rate": 5.6859905792168714e-05,
443
+ "loss": 1.5118,
444
+ "step": 350
445
+ },
446
+ {
447
+ "epoch": 2.59,
448
+ "learning_rate": 4.920454946274896e-05,
449
+ "loss": 1.4238,
450
+ "step": 355
451
+ },
452
+ {
453
+ "epoch": 2.63,
454
+ "learning_rate": 4.180388945025492e-05,
455
+ "loss": 1.7583,
456
+ "step": 360
457
+ },
458
+ {
459
+ "epoch": 2.66,
460
+ "learning_rate": 3.4755109413523463e-05,
461
+ "loss": 1.8061,
462
+ "step": 365
463
+ },
464
+ {
465
+ "epoch": 2.7,
466
+ "learning_rate": 2.8150772209536505e-05,
467
+ "loss": 1.8494,
468
+ "step": 370
469
+ },
470
+ {
471
+ "epoch": 2.74,
472
+ "learning_rate": 2.2077604380601286e-05,
473
+ "loss": 1.687,
474
+ "step": 375
475
+ },
476
+ {
477
+ "epoch": 2.77,
478
+ "learning_rate": 1.6615357282590296e-05,
479
+ "loss": 1.495,
480
+ "step": 380
481
+ },
482
+ {
483
+ "epoch": 2.81,
484
+ "learning_rate": 1.1835759809623895e-05,
485
+ "loss": 1.2369,
486
+ "step": 385
487
+ },
488
+ {
489
+ "epoch": 2.85,
490
+ "learning_rate": 7.801576467760446e-06,
491
+ "loss": 1.8264,
492
+ "step": 390
493
+ },
494
+ {
495
+ "epoch": 2.88,
496
+ "learning_rate": 4.565783166840331e-06,
497
+ "loss": 1.4865,
498
+ "step": 395
499
+ },
500
+ {
501
+ "epoch": 2.92,
502
+ "learning_rate": 2.1708715537833474e-06,
503
+ "loss": 1.805,
504
+ "step": 400
505
+ },
506
+ {
507
+ "epoch": 2.96,
508
+ "learning_rate": 6.482910226644399e-07,
509
+ "loss": 1.5174,
510
+ "step": 405
511
+ },
512
+ {
513
+ "epoch": 2.99,
514
+ "learning_rate": 1.8035728955987417e-08,
515
+ "loss": 1.5703,
516
+ "step": 410
517
+ },
518
+ {
519
+ "epoch": 3.0,
520
+ "eval_loss": 1.8793048858642578,
521
+ "eval_runtime": 8.0763,
522
+ "eval_samples_per_second": 22.411,
523
+ "eval_steps_per_second": 2.848,
524
+ "step": 411
525
+ },
526
+ {
527
+ "epoch": 3.03,
528
+ "learning_rate": 2.883820311639419e-07,
529
+ "loss": 1.4287,
530
+ "step": 415
531
+ },
532
+ {
533
+ "epoch": 3.07,
534
+ "learning_rate": 1.4557798077103288e-06,
535
+ "loss": 1.4968,
536
+ "step": 420
537
+ },
538
+ {
539
+ "epoch": 3.1,
540
+ "learning_rate": 3.5048990762623317e-06,
541
+ "loss": 1.3425,
542
+ "step": 425
543
+ },
544
+ {
545
+ "epoch": 3.14,
546
+ "learning_rate": 6.408831303314137e-06,
547
+ "loss": 1.3711,
548
+ "step": 430
549
+ },
550
+ {
551
+ "epoch": 3.18,
552
+ "learning_rate": 1.012944276047348e-05,
553
+ "loss": 1.3321,
554
+ "step": 435
555
+ },
556
+ {
557
+ "epoch": 3.21,
558
+ "learning_rate": 1.461787528726463e-05,
559
+ "loss": 1.5272,
560
+ "step": 440
561
+ },
562
+ {
563
+ "epoch": 3.25,
564
+ "learning_rate": 1.981518788455224e-05,
565
+ "loss": 1.5268,
566
+ "step": 445
567
+ },
568
+ {
569
+ "epoch": 3.28,
570
+ "learning_rate": 2.5653130713339512e-05,
571
+ "loss": 1.3194,
572
+ "step": 450
573
+ },
574
+ {
575
+ "epoch": 3.32,
576
+ "learning_rate": 3.2055041334975546e-05,
577
+ "loss": 1.7163,
578
+ "step": 455
579
+ },
580
+ {
581
+ "epoch": 3.36,
582
+ "learning_rate": 3.8936851423566864e-05,
583
+ "loss": 1.6848,
584
+ "step": 460
585
+ },
586
+ {
587
+ "epoch": 3.39,
588
+ "learning_rate": 4.620819073068958e-05,
589
+ "loss": 1.3757,
590
+ "step": 465
591
+ },
592
+ {
593
+ "epoch": 3.43,
594
+ "learning_rate": 5.377357380540826e-05,
595
+ "loss": 1.4964,
596
+ "step": 470
597
+ },
598
+ {
599
+ "epoch": 3.47,
600
+ "learning_rate": 6.15336538858851e-05,
601
+ "loss": 1.3765,
602
+ "step": 475
603
+ },
604
+ {
605
+ "epoch": 3.5,
606
+ "learning_rate": 6.938652749677947e-05,
607
+ "loss": 1.4432,
608
+ "step": 480
609
+ },
610
+ {
611
+ "epoch": 3.54,
612
+ "learning_rate": 7.722907262078417e-05,
613
+ "loss": 1.4727,
614
+ "step": 485
615
+ },
616
+ {
617
+ "epoch": 3.58,
618
+ "learning_rate": 8.495830287175848e-05,
619
+ "loss": 1.4478,
620
+ "step": 490
621
+ },
622
+ {
623
+ "epoch": 3.61,
624
+ "learning_rate": 9.247271988678396e-05,
625
+ "loss": 1.5617,
626
+ "step": 495
627
+ },
628
+ {
629
+ "epoch": 3.65,
630
+ "learning_rate": 9.967364617786429e-05,
631
+ "loss": 1.6394,
632
+ "step": 500
633
+ },
634
+ {
635
+ "epoch": 3.69,
636
+ "learning_rate": 0.00010646652094058468,
637
+ "loss": 1.3227,
638
+ "step": 505
639
+ },
640
+ {
641
+ "epoch": 3.72,
642
+ "learning_rate": 0.00011276214180348806,
643
+ "loss": 1.6685,
644
+ "step": 510
645
+ },
646
+ {
647
+ "epoch": 3.76,
648
+ "learning_rate": 0.00011847783621181909,
649
+ "loss": 1.4602,
650
+ "step": 515
651
+ },
652
+ {
653
+ "epoch": 3.8,
654
+ "learning_rate": 0.0001235385470633116,
655
+ "loss": 1.5915,
656
+ "step": 520
657
+ },
658
+ {
659
+ "epoch": 3.83,
660
+ "learning_rate": 0.0001278778183397141,
661
+ "loss": 1.6258,
662
+ "step": 525
663
+ },
664
+ {
665
+ "epoch": 3.87,
666
+ "learning_rate": 0.0001314386677909818,
667
+ "loss": 1.4522,
668
+ "step": 530
669
+ },
670
+ {
671
+ "epoch": 3.91,
672
+ "learning_rate": 0.0001341743352122604,
673
+ "loss": 1.7715,
674
+ "step": 535
675
+ },
676
+ {
677
+ "epoch": 3.94,
678
+ "learning_rate": 0.0001360488964874695,
679
+ "loss": 1.5622,
680
+ "step": 540
681
+ },
682
+ {
683
+ "epoch": 3.98,
684
+ "learning_rate": 0.0001370377353360163,
685
+ "loss": 1.5959,
686
+ "step": 545
687
+ },
688
+ {
689
+ "epoch": 4.0,
690
+ "eval_loss": 1.901490569114685,
691
+ "eval_runtime": 8.0503,
692
+ "eval_samples_per_second": 22.484,
693
+ "eval_steps_per_second": 2.857,
694
+ "step": 548
695
+ },
696
+ {
697
+ "epoch": 4.01,
698
+ "learning_rate": 0.00013712786656777722,
699
+ "loss": 1.2578,
700
+ "step": 550
701
+ },
702
+ {
703
+ "epoch": 4.05,
704
+ "learning_rate": 0.0001363181066014416,
705
+ "loss": 1.2876,
706
+ "step": 555
707
+ },
708
+ {
709
+ "epoch": 4.09,
710
+ "learning_rate": 0.00013461908900701188,
711
+ "loss": 1.3214,
712
+ "step": 560
713
+ },
714
+ {
715
+ "epoch": 4.12,
716
+ "learning_rate": 0.00013205312486836134,
717
+ "loss": 1.3761,
718
+ "step": 565
719
+ },
720
+ {
721
+ "epoch": 4.16,
722
+ "learning_rate": 0.0001286539097995321,
723
+ "loss": 1.4559,
724
+ "step": 570
725
+ },
726
+ {
727
+ "epoch": 4.2,
728
+ "learning_rate": 0.00012446608146216587,
729
+ "loss": 1.1372,
730
+ "step": 575
731
+ },
732
+ {
733
+ "epoch": 4.23,
734
+ "learning_rate": 0.0001195446333946388,
735
+ "loss": 1.1916,
736
+ "step": 580
737
+ },
738
+ {
739
+ "epoch": 4.27,
740
+ "learning_rate": 0.00011395419285035411,
741
+ "loss": 1.3227,
742
+ "step": 585
743
+ },
744
+ {
745
+ "epoch": 4.31,
746
+ "learning_rate": 0.00010776817212844265,
747
+ "loss": 1.3371,
748
+ "step": 590
749
+ },
750
+ {
751
+ "epoch": 4.34,
752
+ "learning_rate": 0.00010106780454139066,
753
+ "loss": 1.2405,
754
+ "step": 595
755
+ },
756
+ {
757
+ "epoch": 4.38,
758
+ "learning_rate": 9.394107767902941e-05,
759
+ "loss": 1.3486,
760
+ "step": 600
761
+ },
762
+ {
763
+ "epoch": 4.42,
764
+ "learning_rate": 8.648157797700773e-05,
765
+ "loss": 1.4013,
766
+ "step": 605
767
+ },
768
+ {
769
+ "epoch": 4.45,
770
+ "learning_rate": 7.878726176258706e-05,
771
+ "loss": 0.9863,
772
+ "step": 610
773
+ },
774
+ {
775
+ "epoch": 4.49,
776
+ "learning_rate": 7.095916891608919e-05,
777
+ "loss": 1.4232,
778
+ "step": 615
779
+ },
780
+ {
781
+ "epoch": 4.53,
782
+ "learning_rate": 6.310009603987893e-05,
783
+ "loss": 1.3065,
784
+ "step": 620
785
+ },
786
+ {
787
+ "epoch": 4.56,
788
+ "learning_rate": 5.531324655850025e-05,
789
+ "loss": 1.4746,
790
+ "step": 625
791
+ },
792
+ {
793
+ "epoch": 4.6,
794
+ "learning_rate": 4.77008754765251e-05,
795
+ "loss": 1.286,
796
+ "step": 630
797
+ },
798
+ {
799
+ "epoch": 4.64,
800
+ "learning_rate": 4.036294659082055e-05,
801
+ "loss": 1.4165,
802
+ "step": 635
803
+ },
804
+ {
805
+ "epoch": 4.67,
806
+ "learning_rate": 3.3395819790395745e-05,
807
+ "loss": 1.3111,
808
+ "step": 640
809
+ },
810
+ {
811
+ "epoch": 4.71,
812
+ "learning_rate": 2.689098568188601e-05,
813
+ "loss": 1.1802,
814
+ "step": 645
815
+ },
816
+ {
817
+ "epoch": 4.74,
818
+ "learning_rate": 2.093386415726427e-05,
819
+ "loss": 1.2933,
820
+ "step": 650
821
+ },
822
+ {
823
+ "epoch": 4.78,
824
+ "learning_rate": 1.5602682680691885e-05,
825
+ "loss": 1.2776,
826
+ "step": 655
827
+ },
828
+ {
829
+ "epoch": 4.82,
830
+ "learning_rate": 1.09674490245759e-05,
831
+ "loss": 1.5263,
832
+ "step": 660
833
+ },
834
+ {
835
+ "epoch": 4.85,
836
+ "learning_rate": 7.089031944609217e-06,
837
+ "loss": 1.3473,
838
+ "step": 665
839
+ },
840
+ {
841
+ "epoch": 4.89,
842
+ "learning_rate": 4.018361866144457e-06,
843
+ "loss": 1.2805,
844
+ "step": 670
845
+ },
846
+ {
847
+ "epoch": 4.93,
848
+ "learning_rate": 1.7957620782939648e-06,
849
+ "loss": 1.2707,
850
+ "step": 675
851
+ },
852
+ {
853
+ "epoch": 4.96,
854
+ "learning_rate": 4.504192183509004e-07,
855
+ "loss": 1.1234,
856
+ "step": 680
857
+ },
858
+ {
859
+ "epoch": 5.0,
860
+ "learning_rate": 0.0,
861
+ "loss": 1.7955,
862
+ "step": 685
863
+ },
864
+ {
865
+ "epoch": 5.0,
866
+ "eval_loss": 1.7918421030044556,
867
+ "eval_runtime": 8.0478,
868
+ "eval_samples_per_second": 22.491,
869
+ "eval_steps_per_second": 2.858,
870
+ "step": 685
871
+ },
872
+ {
873
+ "epoch": 5.04,
874
+ "learning_rate": 4.5041921835088515e-07,
875
+ "loss": 1.0568,
876
+ "step": 690
877
+ },
878
+ {
879
+ "epoch": 5.07,
880
+ "learning_rate": 1.795762078293942e-06,
881
+ "loss": 1.1207,
882
+ "step": 695
883
+ },
884
+ {
885
+ "epoch": 5.11,
886
+ "learning_rate": 4.018361866144427e-06,
887
+ "loss": 1.2954,
888
+ "step": 700
889
+ },
890
+ {
891
+ "epoch": 5.15,
892
+ "learning_rate": 7.0890319446091785e-06,
893
+ "loss": 1.2248,
894
+ "step": 705
895
+ },
896
+ {
897
+ "epoch": 5.18,
898
+ "learning_rate": 1.0967449024575855e-05,
899
+ "loss": 1.2091,
900
+ "step": 710
901
+ },
902
+ {
903
+ "epoch": 5.22,
904
+ "learning_rate": 1.5602682680691756e-05,
905
+ "loss": 1.2467,
906
+ "step": 715
907
+ },
908
+ {
909
+ "epoch": 5.26,
910
+ "learning_rate": 2.093386415726413e-05,
911
+ "loss": 1.1092,
912
+ "step": 720
913
+ },
914
+ {
915
+ "epoch": 5.29,
916
+ "learning_rate": 2.6890985681885842e-05,
917
+ "loss": 1.2179,
918
+ "step": 725
919
+ },
920
+ {
921
+ "epoch": 5.33,
922
+ "learning_rate": 3.339581979039557e-05,
923
+ "loss": 1.0797,
924
+ "step": 730
925
+ },
926
+ {
927
+ "epoch": 5.36,
928
+ "learning_rate": 4.036294659082037e-05,
929
+ "loss": 1.2595,
930
+ "step": 735
931
+ },
932
+ {
933
+ "epoch": 5.4,
934
+ "learning_rate": 4.77008754765249e-05,
935
+ "loss": 1.2369,
936
+ "step": 740
937
+ },
938
+ {
939
+ "epoch": 5.44,
940
+ "learning_rate": 5.5313246558500285e-05,
941
+ "loss": 1.2623,
942
+ "step": 745
943
+ },
944
+ {
945
+ "epoch": 5.47,
946
+ "learning_rate": 6.310009603987897e-05,
947
+ "loss": 1.2695,
948
+ "step": 750
949
+ },
950
+ {
951
+ "epoch": 5.51,
952
+ "learning_rate": 7.095916891608921e-05,
953
+ "loss": 1.1535,
954
+ "step": 755
955
+ },
956
+ {
957
+ "epoch": 5.55,
958
+ "learning_rate": 7.87872617625871e-05,
959
+ "loss": 1.0823,
960
+ "step": 760
961
+ },
962
+ {
963
+ "epoch": 5.58,
964
+ "learning_rate": 8.648157797700775e-05,
965
+ "loss": 1.0971,
966
+ "step": 765
967
+ },
968
+ {
969
+ "epoch": 5.62,
970
+ "learning_rate": 9.394107767902946e-05,
971
+ "loss": 0.978,
972
+ "step": 770
973
+ },
974
+ {
975
+ "epoch": 5.66,
976
+ "learning_rate": 0.00010106780454139057,
977
+ "loss": 1.2448,
978
+ "step": 775
979
+ },
980
+ {
981
+ "epoch": 5.69,
982
+ "learning_rate": 0.00010776817212844258,
983
+ "loss": 1.088,
984
+ "step": 780
985
+ },
986
+ {
987
+ "epoch": 5.73,
988
+ "learning_rate": 0.00011395419285035404,
989
+ "loss": 1.1032,
990
+ "step": 785
991
+ },
992
+ {
993
+ "epoch": 5.77,
994
+ "learning_rate": 0.00011954463339463875,
995
+ "loss": 1.1252,
996
+ "step": 790
997
+ },
998
+ {
999
+ "epoch": 5.8,
1000
+ "learning_rate": 0.0001244660814621658,
1001
+ "loss": 1.2392,
1002
+ "step": 795
1003
+ },
1004
+ {
1005
+ "epoch": 5.84,
1006
+ "learning_rate": 0.00012865390979953207,
1007
+ "loss": 1.1824,
1008
+ "step": 800
1009
+ },
1010
+ {
1011
+ "epoch": 5.88,
1012
+ "learning_rate": 0.00013205312486836131,
1013
+ "loss": 1.1355,
1014
+ "step": 805
1015
+ },
1016
+ {
1017
+ "epoch": 5.91,
1018
+ "learning_rate": 0.00013461908900701185,
1019
+ "loss": 1.1338,
1020
+ "step": 810
1021
+ },
1022
+ {
1023
+ "epoch": 5.95,
1024
+ "learning_rate": 0.00013631810660144153,
1025
+ "loss": 1.3598,
1026
+ "step": 815
1027
+ },
1028
+ {
1029
+ "epoch": 5.99,
1030
+ "learning_rate": 0.0001371278665677772,
1031
+ "loss": 1.3885,
1032
+ "step": 820
1033
+ },
1034
+ {
1035
+ "epoch": 6.0,
1036
+ "eval_loss": 1.7604948282241821,
1037
+ "eval_runtime": 8.058,
1038
+ "eval_samples_per_second": 22.462,
1039
+ "eval_steps_per_second": 2.854,
1040
+ "step": 822
1041
+ }
1042
+ ],
1043
+ "max_steps": 2740,
1044
+ "num_train_epochs": 20,
1045
+ "total_flos": 855992696832000.0,
1046
+ "trial_name": null,
1047
+ "trial_params": null
1048
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e887f9fc06cce3b63f585b7c5b8b706b54fff3e04c1880ad0fa1b0ea356cea0
3
+ size 2671
vocab.json ADDED
The diff for this file is too large to render. See raw diff